From 2d985f8c6b91b5007a16e640bb9c038c5fb2839b Mon Sep 17 00:00:00 2001 From: Eric Biggers Date: Fri, 26 Aug 2022 23:58:45 -0700 Subject: vfs: support STATX_DIOALIGN on block devices Add support for STATX_DIOALIGN to block devices, so that direct I/O alignment restrictions are exposed to userspace in a generic way. Note that this breaks the tradition of stat operating only on the block device node, not the block device itself. However, it was felt that doing this is preferable, in order to make the interface useful and avoid needing separate interfaces for regular files and block devices. Signed-off-by: Eric Biggers Reviewed-by: Christoph Hellwig Reviewed-by: Christian Brauner (Microsoft) Link: https://lore.kernel.org/r/20220827065851.135710-3-ebiggers@kernel.org --- include/linux/blkdev.h | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'include/linux/blkdev.h') diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index 84b13fdd34a7..8038c5fbde40 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -1498,6 +1498,7 @@ int sync_blockdev(struct block_device *bdev); int sync_blockdev_range(struct block_device *bdev, loff_t lstart, loff_t lend); int sync_blockdev_nowait(struct block_device *bdev); void sync_bdevs(bool wait); +void bdev_statx_dioalign(struct inode *inode, struct kstat *stat); void printk_all_partitions(void); #else static inline void invalidate_bdev(struct block_device *bdev) @@ -1514,6 +1515,9 @@ static inline int sync_blockdev_nowait(struct block_device *bdev) static inline void sync_bdevs(bool wait) { } +static inline void bdev_statx_dioalign(struct inode *inode, struct kstat *stat) +{ +} static inline void printk_all_partitions(void) { } -- cgit v1.2.3 From 568ec936bf1384fc15873908c96a9aeb62536edb Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Tue, 27 Sep 2022 09:58:15 +0200 Subject: block: replace blk_queue_nowait with bdev_nowait Replace blk_queue_nowait with a bdev_nowait helpers that takes the block_device given that the I/O submission path should not have to look into the request_queue. Signed-off-by: Christoph Hellwig Reviewed-by: Pankaj Raghav Link: https://lore.kernel.org/r/20220927075815.269694-1-hch@lst.de Signed-off-by: Jens Axboe --- include/linux/blkdev.h | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) (limited to 'include/linux/blkdev.h') diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index 84b13fdd34a7..4750772ef228 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -618,7 +618,6 @@ bool blk_queue_flag_test_and_set(unsigned int flag, struct request_queue *q); #define blk_queue_quiesced(q) test_bit(QUEUE_FLAG_QUIESCED, &(q)->queue_flags) #define blk_queue_pm_only(q) atomic_read(&(q)->pm_only) #define blk_queue_registered(q) test_bit(QUEUE_FLAG_REGISTERED, &(q)->queue_flags) -#define blk_queue_nowait(q) test_bit(QUEUE_FLAG_NOWAIT, &(q)->queue_flags) #define blk_queue_sq_sched(q) test_bit(QUEUE_FLAG_SQ_SCHED, &(q)->queue_flags) extern void blk_set_pm_only(struct request_queue *q); @@ -1280,6 +1279,11 @@ static inline bool bdev_fua(struct block_device *bdev) return test_bit(QUEUE_FLAG_FUA, &bdev_get_queue(bdev)->queue_flags); } +static inline bool bdev_nowait(struct block_device *bdev) +{ + return test_bit(QUEUE_FLAG_NOWAIT, &bdev_get_queue(bdev)->queue_flags); +} + static inline enum blk_zoned_model bdev_zoned_model(struct block_device *bdev) { struct request_queue *q = bdev_get_queue(bdev); -- cgit v1.2.3 From 8cafdb5ab94cda3ebb0975be16e2d564a05132ea Mon Sep 17 00:00:00 2001 From: Pankaj Raghav Date: Thu, 29 Sep 2022 09:47:44 +0200 Subject: block: adapt blk_mq_plug() to not plug for writes that require a zone lock The current implementation of blk_mq_plug() disables plugging for all operations that involves a transfer to the device as we just check if the last bit in op_is_write() function. Modify blk_mq_plug() to disable plugging only for REQ_OP_WRITE and REQ_OP_WRITE_ZEROS as they might require a zone lock. Suggested-by: Christoph Hellwig Suggested-by: Damien Le Moal Reviewed-by: Christoph Hellwig Signed-off-by: Pankaj Raghav Reviewed-by: Damien Le Moal Reviewed-by: Johannes Thumshirn Link: https://lore.kernel.org/r/20220929074745.103073-2-p.raghav@samsung.com Signed-off-by: Jens Axboe --- include/linux/blkdev.h | 9 +++++++++ 1 file changed, 9 insertions(+) (limited to 'include/linux/blkdev.h') diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index 4750772ef228..49373d002631 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -1304,6 +1304,15 @@ static inline bool bdev_is_zoned(struct block_device *bdev) return false; } +static inline bool bdev_op_is_zoned_write(struct block_device *bdev, + blk_opf_t op) +{ + if (!bdev_is_zoned(bdev)) + return false; + + return op == REQ_OP_WRITE || op == REQ_OP_WRITE_ZEROES; +} + static inline sector_t bdev_zone_sectors(struct block_device *bdev) { struct request_queue *q = bdev_get_queue(bdev); -- cgit v1.2.3 From ca5eebda3e1c1a58a1c5a337da393ed6734593e3 Mon Sep 17 00:00:00 2001 From: Brian Foster Date: Mon, 3 Oct 2022 09:35:34 -0400 Subject: block: avoid sign extend problem with default queue flags mask request_queue->queue_flags is unsigned long, which is 8-bytes on 64-bit architectures. Most queue flag modifications occur through bit field helpers, but default flags can be logically OR'd via the QUEUE_FLAG_MQ_DEFAULT mask. If this mask happens to include bit 31, the assignment can sign extend the field and set all upper 32 bits. This exact problem has been observed on a downstream kernel that happens to use bit 31 for QUEUE_FLAG_NOWAIT. This is not an immediate problem for current upstream because bit 31 is not included in the default flag assignment (and is not used at all, actually). Regardless, fix up the QUEUE_FLAG_MQ_DEFAULT mask definition to avoid the landmine in the future. Signed-off-by: Brian Foster Reviewed-by: Christoph Hellwig Link: https://lore.kernel.org/r/20221003133534.1075582-1-bfoster@redhat.com Signed-off-by: Jens Axboe --- include/linux/blkdev.h | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'include/linux/blkdev.h') diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index 49373d002631..e4fb47753177 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -580,9 +580,9 @@ struct request_queue { #define QUEUE_FLAG_NOWAIT 29 /* device supports NOWAIT */ #define QUEUE_FLAG_SQ_SCHED 30 /* single queue style io dispatch */ -#define QUEUE_FLAG_MQ_DEFAULT ((1 << QUEUE_FLAG_IO_STAT) | \ - (1 << QUEUE_FLAG_SAME_COMP) | \ - (1 << QUEUE_FLAG_NOWAIT)) +#define QUEUE_FLAG_MQ_DEFAULT ((1UL << QUEUE_FLAG_IO_STAT) | \ + (1UL << QUEUE_FLAG_SAME_COMP) | \ + (1UL << QUEUE_FLAG_NOWAIT)) void blk_queue_flag_set(unsigned int flag, struct request_queue *q); void blk_queue_flag_clear(unsigned int flag, struct request_queue *q); -- cgit v1.2.3